With the westernization of the life style and the increasing number of aged people, ischemic heart diseases and other pathology of heart and blood vessels are increasing year after year. In particular, increase of vascular occlusive diseases such as myocardial infarction, cerebral thrombosis, pulmonary embolism, and peripheral arterial and venous occlusive disease is increasing each year and treatment of such diseases has become a serious social issue. In the treatment and prevention of the thrombosis, anticoagulation therapy has been playing an important role in internal medicine together with anti-platelet therapy and thrombolytic therapy. For the treatment and prevention of thrombosis, safety that permits long-term drug administration and the development of a positive and appropriate anticoagulant activity are essential. Heretofore, anticoagulants such as warfarin and heparin have been used in order to prevent and treat thrombosis due to hypercoagulability. However, such use of the anticoagulants has been pointed to be associated with various demerits including the risk of bleeding and interactions with other drugs. Warfarin is extensively used in the world as the solo peroral anticoagulant. However, due to its characteristics based on the mechanism of action, the concentration range for the development of efficacy is narrow and yet it takes long to develop efficacy and the half-life in blood is as long as 36 hours; what is more, for several reasons such as the great individual difference of effective dose, it is difficult to control the anticoagulability of warfarin (N. Eng. J. Med. 324 (26) 1865-1875, 1991) and frequent monitoring is required in order to prevent bleeding as a side effect. In addition, warfarin also has many other side effects such as nausea, vomiting, diarrhea, and alopecia; thus, warfarin is a drug that involves considerable difficulty in clinical use. On the other hand, heparin is extensively used in the world as an intravenously administrable anticoagulant. However, since it is a direct inhibitor of thrombin, heparin has a high risk of bleeding and needs as frequent monitoring as warfarin; what is more, due to its characteristics based on the mechanism of action, adequate coagulation inhibiting effect is not expected at a lowered antithrombin III level; thus, heparin is a drug that involves considerable difficulty in clinical use. In view of such situation, improved anticoagulants have been desired that has none of the defects inherent in warfarin and heparin.
The blood coagulation cascade is a chain reaction involving limited protein hydrolysis triggered by activation of the extrinsic coagulation cascade or the intrinsic coagulation cascade, and once the cascade is activated, the reaction is amplifies like an avalanche. Since the final stage of the blood coagulation cascade is thrombin-mediated conversion of fibrinogen to fibrin, efforts have recently been made to develop thrombin inhibitors; however, drugs that directly inhibit thrombin are known to increase the risk of bleeding. In addition, they have low bioavailability in oral administration; therefore no thrombin inhibitor, which can be orally administered, has been introduced into market.
FXa is a key enzyme, which is located in the upstream of the thrombin in the coagulation cascade, and also at the point of convergence between the extrinsic and the intrinsic coagulation cascade. One molecule of FXa is known to produce about a hundred molecules of thrombin per minute. Therefore, an FXa inhibitor can potentially inhibit the coagulation cascade more efficiently than a thrombin inhibitor (Thrombosis Research, vol. 19, pages 339-349, 1980; Mebio vol. 14, No. 8, 1997).
Compounds that exhibit FXa inhibiting actions have been disclosed in several patents, among which Japanese Patent Application Laid-Opened No.208946/1993 and WO96/16940 disclose aromatic amidine derivatives, and in particular, amidinonaphthyl derivatives, and WO97/38984 and the like disclose cyclic urea compounds having an aminodiphenyl group. However, these compounds are still in the process of development and none have been commercialized to date.
These compounds also suffer from low bioavailability in oral administration, and there is good room of improvement in separating the thrombin inhibitory action and the trypsin inhibitory action from the FXa inhibitory action. In addition, there is apprehension that these compounds are associated with decrease of blood pressure, respiratory insufficiency, and other side effects induced by the amidino group.
With regard to the compound of Japanese Patent Application Laid-Opened No. 208946/1993, use of this compound as a preventive and therapeutic agent for influenza virus is disclosed. The activity of this agent to inhibit the influenza virus propagation is based on the FXa inhibitory action.
Compounds having an aminoheterocyclic group typified by 1-(4-pyridyl)piperidin-4-yl group can be used as FXa inhibitor; for example, disclosed in prior art references including WO96/10022, WO97/28129, WO97/29104, WO98/21188, WO98/54164, WO99/06371, and WO99/09027.
These compounds have been developed for the purpose of providing an FXa inhibitor, which is effective in oral absorption. However, low molecular weight FXa inhibitors are still under development and no drug of low molecular weight FXa inhibitors has been commercialized.
In the development of pharmaceutical products, the desired pharmacological activity is not the sole requirement. Another requirement is that strict criteria be met in various aspects including absorption, distribution, metabolism and excretion, and the like. For example, the drugs are required to pass various examinations for drug interaction, desensitization or tolerance, absorption from digestive tract in the oral administration, transfer rate to small intestine, absorption rate and first pass effect, organ barrier, protein binding, induction of drug metabolizing enzyme, excretion pathway and clearance from body, administration method (site, method, and purpose of administration), and the like, and a drug meeting all such requirements are seldom discovered.
The anticoagulants also share such general challenge of the drug development.
In the case of the FXa inhibitor, circumvention of the problem of the side effects associated with the oral administration of the warfarin as well as risk of bleeding based on the thrombin inhibition as found in the case of heparin whose administration is only accomplished by intravenous injection is required.
When the FXa inhibitor is constructed by molecular designing method, condition of the binding between the FXa and the FXa inhibitor has great significance. In the three-dimensional structure of FXa, active site of the FXa is formed in the structure characteristic to a chymotrypsin-like serine protease.
The active center of a serine protease is formed from a plurality of pockets called subsites, and it is known that substantially all of the inhibitors which do not form covalent bond with Ser195 residue bind to these pockets. Among such pockets, S1 pocket is believed to be the most important in the serine protease in the binding with the substrate, or in the development of the substrate selectivity.
S1 site is also believed to be the most important in the serine protease inhibitor for the development of the inhibitory activity and the enzyme selectivity. The residue which is generally believed to be the most important for the substrate specificity in the S1 pocket is the residue corresponding to chymotrypsin No. 189, and FXa has Asp (Asp189) as this residue and the inside of the pocket is believed to be negatively charged.
However, this is also the case in serine proteases other than the FXa, namely, in trypsin, thrombin, protein C, tissue plasminogen activator, and the like, and such resemblance is one cause of difficulty in enzyme specificity in FXa inhibitor. The substrate specificity of the entire enzyme is determined by structural difference of the subsite such as S3 pocket in addition to the structural difference of S1, and designing of an inhibitor selective for FXa can be accomplished by using such structural difference.
With regard to the binding state between the FXa and the FXa inhibitor, there has been only the report for specific compounds (DX-9065a and FX-2212a) by X-ray crystallographic analysis.
In DX-9065a, amidino group is bound to S1 pocket of the FXa, and in particular, Asp189 and the amidino group are firmly bound to each other by electrostatic interaction and hydrogen bonds. This is the binding manner commonly known in trypsin inhibitors and thrombin inhibitors.
It has also been found out for FX-2212a that the amidino group is bound to the S1 pocket of the FXa by the similar known binding manner (J. Biol. Chem. 1996 Nov. 22; 271(47):29988-92 Brandstetter H. et al., Proc. Natl. Acad. Sci. USA (1998) Jun. 9; 95(12):6630-5 Kamata K. et al.).
However, DX-9065a and FX-2212a are insufficient in their efficacy in oral administration, and there was also apprehension for the side effects induced by amidino group and guanidino group. In the meanwhile, it has not been even found out for the FXa inhibitors having other structure whether such X-ray crystallographic analysis is possible, and the binding state with the FXa was not at all found out.
Accordingly, despite the recognition of the usefulness of the amidino group and the guanidino group, there has been no indication as to what structure should be focused in the effort of searching or developing the FXa inhibitor of different type wherein such defects have been improved since no clue has been provided from the information on interaction based on the crystal structure data of such complex of the FXa with known FXa inhibitory compound, and the structural clue has been sincerely awaited.